Why a Fluoroquinolone Antibiotic Can Damage a Tendon Months After You Stop Taking It

I've written more prescriptions for ciprofloxacin and levofloxacin than I can count, usually for a urinary or respiratory infection that needed knocking down quickly, and most patients take the course, recover, and never think about it again. Every so often one comes back weeks or even months later with a complaint that doesn't seem to fit, a deep ache behind the ankle, a pop during an easy jog, an Achilles that gave out doing something the person had done ten thousand times before. The infection cleared long ago, and the drug left the bloodstream within days, so why would a tendon fail now?

The FDA placed a black box warning on the fluoroquinolone class for this exact problem, and the warning confirms that the injury is real. What it doesn't explain is why the danger lingers so far past the final pill, and the answer runs through two places at once, the tendon itself and, I'll argue, the gut.

The damage outlasts the drug

Fluoroquinolone tendon injury is well documented and a little strange in its timing. The Achilles takes the worst of it, though the rotator cuff and other tendons can be hit, and the trouble often surfaces within the first week or two of treatment. A meaningful share of cases, though, announce themselves a month or more after the last dose, which is the part that makes no intuitive sense. In a 2019 systematic review and meta-analysis in the European Journal of Clinical Pharmacology, Alves and colleagues pooled the observational data and found that fluoroquinolone use roughly doubled the odds of tendon injury, with the risk climbing further in people over 60 and in anyone also taking a corticosteroid [1]. Lewis and Cook, reviewing the literature for athletes and clinicians in the Journal of Athletic Training in 2014, described the same delayed, unpredictable pattern and the lingering vulnerability of the tissue after the course ends [2].

Two honest caveats belong here. The absolute risk for any single person taking a short course is low, so this is a reason for awareness rather than alarm, and these drugs clear serious infections, so no one should refuse a necessary antibiotic on the strength of a blog post. The question worth chasing is mechanistic, why a drug that leaves the body within days can set up an injury that surfaces weeks later.

What the antibiotic does to the tendon

The first part of the answer is direct chemical damage during treatment, and it happens for three reasons at once.

Fluoroquinolones bind metal ions. The same chemistry that makes you take these pills away from antacids and supplements, shown for norfloxacin by Wallis and colleagues in a 1996 study in the Journal of Pharmaceutical Sciences, comes from the drug forming tight complexes with divalent and trivalent cations like magnesium, calcium, iron, and aluminum [3]. That chelation, the trapping of a metal ion inside a larger molecule, matters because tendon cells lean on those same metals to do their work.

At the same time, the drug pushes tendon cells, called tenocytes, toward tearing down their own scaffolding. Tenocytes maintain the collagen framework that gives a tendon its strength, balancing construction against demolition with enzymes called matrix metalloproteinases, or MMPs, which cut collagen and are normally held in check by the cells' own inhibitors. Ciprofloxacin tips that balance toward demolition. Tsai and colleagues, in a 2011 study in the Journal of Orthopaedic Research, showed that ciprofloxacin drove tendon cells to express more MMP-2 and to degrade more type I collagen, the main structural protein of the tissue [4]. Williams and colleagues, back in 2000 in the American Journal of Sports Medicine, had already found that ciprofloxacin slowed tendon fibroblasts and cut their synthesis of collagen and matrix [5]. More breakdown paired with less repair leaves the tendon in what researchers call a net degradative state.

The third reason is oxidative stress, the cellular damage done by reactive oxygen species, unstable molecules that injure proteins, membranes, and DNA. Pouzaud and colleagues, in a 2004 study in the Journal of Pharmacology and Experimental Therapeutics, traced fluoroquinolone toxicity in tendon cells to exactly this kind of oxidative injury [6]. Lowes and colleagues sharpened the picture in 2009 in Free Radical Research, showing that fluoroquinolones damaged the mitochondria, the cell's energy plants, inside human Achilles tendon cells, and that a mitochondria-targeted antioxidant called MitoQ prevented much of it [7]. By the time the prescription is finished, the tissue may already be quietly weakened.

The magnesium clue

One of those three threads, the metal binding, turns out to be more than a footnote, and the evidence for it is some of the most elegant in this whole story. If chelating magnesium were part of how these drugs hurt connective tissue, then simply starving an animal of magnesium ought to reproduce the damage, and it does. Stahlmann and colleagues, in a 1995 study in Antimicrobial Agents and Chemotherapy, fed juvenile rats a magnesium-deficient diet and produced joint cartilage lesions that were, in their words, identical to the arthropathy that quinolone antibiotics cause [8]. Shakibaei and colleagues took the next step in 2000 in the same journal, showing that rats given ofloxacin while fed a magnesium-poor diet developed Achilles tendon damage visible under the electron microscope, worse than either insult alone [9].

This is the hinge of the argument. A mineral shortfall and the drug do the same kind of harm to the same tissue, and together they do more. Tendon integrity depends on having enough of the right minerals on hand, and a fluoroquinolone undercuts that supply at the chemical level. So where do those minerals come from, and what happens to that supply line after a course of antibiotics?

The antibiotic also empties your gut

Fluoroquinolones are broad-spectrum, which means they don't distinguish between the bacterium causing your infection and the trillions of harmless and helpful bacteria living in your colon. A single course can clear out a large fraction of that community, and the recovery is neither quick nor guaranteed. Dethlefsen and Relman, in a 2011 study in the Proceedings of the National Academy of Sciences, gave healthy volunteers two five-day courses of ciprofloxacin and watched the gut community shift within a couple of days, recover only partway over the following weeks, and in places settle into a new state that still differed from where it started, with the pattern of loss and return differing from one person to the next [10]. Some microbiomes bounce back in weeks, others take six months or longer, and a few never fully return to baseline.

I think that variability is the missing piece, because it lines up with the clinical reality that some people take these drugs without trouble while others rupture a tendon weeks after finishing. The reason the gut matters to a tendon has to do with minerals.

Gut bacteria help you absorb minerals

Your body can't manufacture copper, zinc, or magnesium, so every atom you own came in through the gut wall, and how much you absorb is shaped in part by the bacteria living there. They help in a few concrete ways.

When gut bacteria ferment the fiber you eat, they produce short-chain fatty acids, butyrate among them, the same fuel the colon lining runs on (I've written about butyrate in our post on the microbiome and colon cancer). Those acids lower the pH inside the colon, which keeps minerals dissolved and easier to pull across the gut wall. Scholz-Ahrens and colleagues, in a 2007 review in the Journal of Nutrition, gathered the animal and human evidence that fermentable fibers and the bacteria feeding on them raise the absorption of calcium and magnesium, partly through that drop in pH [11].

Bacteria also unlock minerals that plant foods otherwise hold hostage. Whole grains, legumes, nuts, and seeds carry phytate, a storage compound that clamps onto minerals and blocks their uptake, and certain lactic acid bacteria, including Lactobacillus and Bifidobacterium species, make an enzyme called phytase that pries phytate apart and frees what it was holding. García-Mantrana and colleagues, in a 2015 study in Plant Foods for Human Nutrition, showed that lactic acid bacteria carrying these phytases cut the phytate in a soy drink and improved its mineral availability [12]. A gut stripped of those bacteria leaves more phytate intact, so more of the copper and zinc on your plate passes straight through you. The same butyrate that lowers colonic pH also helps hold the gut lining's tight junctions closed, so a microbiome in disarray tends to absorb minerals less well across the board. I've made this same case about the skin, which is built from the very minerals an inflamed gut fails to let in.

What a healing tendon is built from

Now put the tendon's repair list next to that absorption problem. A tendon trying to rebuild collagen needs copper, because copper is the required cofactor for lysyl oxidase, the enzyme that cross-links collagen and elastin fibers into something that can bear load. Rucker and colleagues, in a 1998 paper in the American Journal of Clinical Nutrition, detailed how copper feeds lysyl oxidase to forge those cross-links, and how a copper shortfall leaves the connective tissue weak [13]. New collagen laid down without enough copper stays slack, because the cross-links that lock it into a load-bearing cable never properly form.

It needs zinc, which drives collagen synthesis and sits in the catalytic core of the very MMPs that remodel a tendon. Lansdown and colleagues, reviewing zinc in wound healing in Wound Repair and Regeneration in 2007, laid out how central zinc is to building and remodeling connective tissue, and how a deficiency stalls repair [14]. It also needs magnesium, the same mineral the drug chelates and the one whose absence reproduced the injury in the animal work above.

So in the weeks after a course of fluoroquinolones, the tendon is calling for copper, zinc, and magnesium at the precise moment its owner's gut has been stripped of the bacteria that make those minerals easy to absorb.

My read of how the pieces fit

Here is where I have to be clear about what is established and what is mine. The pieces above are each well supported on their own. Fluoroquinolones damage tendon cells directly, magnesium status changes how much harm they do, the drugs disrupt the gut microbiome for months, the microbiome shapes mineral absorption, and tendons rebuild using copper, zinc, and magnesium. Stringing them into a single causal chain, the idea that an antibiotic-driven gut imbalance starves a healing tendon of minerals and is part of why the risk persists, is my hypothesis, not a proven fact. No one has tracked tendon mineral content and microbiome recovery in the same patients after a course, so I hold this as a strong lead rather than a settled mechanism.

With that said, I find the model hard to ignore, because it explains the timing that nothing else does. Think of it as three overlapping stages. The first is the direct hit, the MMP surge, the oxidative damage, and the metal chelation that weaken the tendon while you're still on the drug. The second is a repair failure, since the same chelation and mineral interference leave the surviving cells unable to lay down sound collagen. The third, and the one I think explains the long tail of risk, is malabsorption, because a gut still recovering from the antibiotic can't efficiently deliver the copper, zinc, and magnesium that repair demands, and that stage can run for weeks to months after the last pill. The person whose microbiome rebounds in two weeks may sail through, while the person whose gut takes six months to recover spends six months trying to rebuild a tendon on a mineral budget that keeps coming up short.

Keeping it honest

None of this means fluoroquinolones are poison or that you should turn one down when you genuinely need it. The absolute risk of tendon injury from a single course is low, most people heal their gut and their tendons without incident, and the mineral-malabsorption chain I've drawn is a reasoned hypothesis built from solid parts, not a result lifted whole from a trial. The FDA already advises reserving these drugs for infections without good alternatives, which is a sensible default whatever the mechanism turns out to be. Treat what follows as low-risk, supportive measures, not as a reason to fear a drug your physician has good cause to prescribe.

What to do

If you're taking a fluoroquinolone now, a few things help, and the one that matters most is not stopping a prescribed course on your own.

• Ask your physician whether a fluoroquinolone is the right choice, since for many routine infections there's an equally good alternative without this risk. If you need it, take it.
• Don't take mineral supplements, copper, zinc, magnesium, calcium, or iron, within about two hours of your dose, because they bind the antibiotic in your gut and blunt its effect. Spacing them well away from the dose, with your physician's okay, is how you get minerals without sabotaging the drug.
• Go easy on the tendons. Avoid heavy resistance work and high-impact running during the course and for several weeks after, since loading a chemically weakened tendon is what turns a vulnerability into a tear.
• Report any tendon pain, swelling, or stiffness right away, since even mild symptoms can come before a rupture, and stopping the drug early, with your prescriber, is the standard response.

If you've recently finished a course, shift the focus to rebuilding the gut and refilling the minerals.

• Feed the microbiome. Eat fermented foods like yogurt, kefir, sauerkraut, and kimchi, give your bacteria a range of fibers to ferment, and lean on whole foods over processed ones (our piece on artificial sweeteners covers what to stop feeding it).
• Get your minerals from food, where they arrive in absorbable form and reasonable balance. Shellfish, liver, and nuts carry copper, oysters and red meat carry zinc, and seeds, leafy greens, and legumes carry magnesium (our guides to minerals in animal foods and zinc and copper balance lay out the specifics). Food beats supplements here, both because absorption is better and because it's hard to throw copper and zinc out of balance with a plate of real food.
• Bring the tendons back slowly. Tendons remodel over months, not weeks, so ramp up load gradually and give a post-antibiotic tendon extra runway before you ask it to sprint or lift heavy.

The damage these drugs do to a tendon outlasts the prescription, and I think part of the reason is that the same course empties the gut of the bacteria you need to absorb the minerals a tendon rebuilds with. The direct toxicity is established, the microbiome disruption is established, and the mineral dependence of repair is established. The line connecting them is my read of how those facts fit together, offered as a lead worth taking seriously, and the steps it points to, protect the tendon, support the gut, and eat for your minerals, cost little and carry almost no downside whether or not the full hypothesis holds.

See how our metabolic program works

References

1. Alves C, Mendes D, Marques FB. Fluoroquinolones and the risk of tendon injury: a systematic review and meta-analysis. Eur J Clin Pharmacol. 2019. PMID: 31270563
2. Lewis T, Cook J. Fluoroquinolones and tendinopathy: a guide for athletes and sports clinicians and a systematic review of the literature. J Athl Train. 2014. PMID: 24762232
3. Wallis SC, Charles BG, Gahan LR, et al. Interaction of norfloxacin with divalent and trivalent pharmaceutical cations. In vitro complexation and in vivo pharmacokinetic studies in the dog. J Pharm Sci. 1996. PMID: 8863267
4. Tsai WC, Hsu CC, Chen CP, et al. Ciprofloxacin up-regulates tendon cells to express matrix metalloproteinase-2 with degradation of type I collagen. J Orthop Res. 2011. PMID: 20602464
5. Williams RJ 3rd, Attia E, Wickiewicz TL, et al. The effect of ciprofloxacin on tendon, paratenon, and capsular fibroblast metabolism. Am J Sports Med. 2000. PMID: 10843129
6. Pouzaud F, Bernard-Beaubois K, Thevenin M, et al. In vitro discrimination of fluoroquinolones toxicity on tendon cells: involvement of oxidative stress. J Pharmacol Exp Ther. 2004. PMID: 14569066
7. Lowes DA, Wallace C, Murphy MP, et al. The mitochondria targeted antioxidant MitoQ protects against fluoroquinolone-induced oxidative stress and mitochondrial membrane damage in human Achilles tendon cells. Free Radic Res. 2009. PMID: 19235604
8. Stahlmann R, Förster C, Shakibaei M, et al. Magnesium deficiency induces joint cartilage lesions in juvenile rats which are identical to quinolone-induced arthropathy. Antimicrob Agents Chemother. 1995. PMID: 8540708
9. Shakibaei M, Pfister K, Schwabe R, et al. Ultrastructure of Achilles tendons of rats treated with ofloxacin and fed a normal or magnesium-deficient diet. Antimicrob Agents Chemother. 2000. PMID: 10639347
10. Dethlefsen L, Relman DA. Incomplete recovery and individualized responses of the human distal gut microbiota to repeated antibiotic perturbation. Proc Natl Acad Sci U S A. 2011. PMID: 20847294
11. Scholz-Ahrens KE, Ade P, Marten B, et al. Prebiotics, probiotics, and synbiotics affect mineral absorption, bone mineral content, and bone structure. J Nutr. 2007. PMID: 17311984
12. García-Mantrana I, Monedero V, Haros M. Reduction of Phytate in Soy Drink by Fermentation with Lactobacillus casei Expressing Phytases From Bifidobacteria. Plant Foods Hum Nutr. 2015. PMID: 26003176
13. Rucker RB, Kosonen T, Clegg MS, et al. Copper, lysyl oxidase, and extracellular matrix protein cross-linking. Am J Clin Nutr. 1998. PMID: 9587142
14. Lansdown AB, Mirastschijski U, Stubbs N, et al. Zinc in wound healing: theoretical, experimental, and clinical aspects. Wound Repair Regen. 2007. PMID: 17244314